Moti Ben-Ari interview: Selected quotes

I still think mathematics is a good thing to major in because I found later on, that when I started working in industry, that anytime something came up, it was really just a bunch of equations. And when you have the mathematical background and you’re not afraid of going and looking at the literature, then you can get into everything much faster than you can otherwise.

[W]hen I see entire companies running around themselves for three or four weeks trying to find a bug and I see this alternate approach, where this is a compilation error, then I became very pro-extreme-Pascal/Ada, these sort of things. And I probably would have made more money if I’d been an expert in C and C++, but I just can’t do that. In fact, the last industrial work I ever did, I personally had to debug for two or three days because of an inconsistency in the declaration, the definition of a C function.

But the problem with constructivism […] I started getting more into what this really is and I started becoming part of those philosophers […] of science and education, who say, “Constructivism is either meaningless or trivial.” And, well, it’s either meaningless in the sense that it’s related to philosophical doctrines like idealism, non-realistic philosophical doctrines that nobody really accepts unless you’re a post-modernist or something like that. And its pedagogical claims, that students should be active in constructing knowledge, is something that is almost trivial. Everybody knows that if they do exercises in labs and talk together it’s better than just hearing the lecture. So I haven’t really pursued this. The only thing that this did lead to was an interest in the philosophy and history of science and science teaching and I actually wrote a book on this.

And I think if I had a piece of advice to give to students that if you really want to do something fun during your professional career, the best thing to do is learn computer science and something else. For example: computer science and physics, or computer science and biology, or computer science and finance. Because then you’re in the really best position. On the one hand, you’re not just playing around with tools, which is okay, but it’s not going to get you very far. But once you start understanding the business you’re in, then you very quickly get to very interesting positions where you understand what’s being done and you have to come up with the computer solutions to do this. So if there’s really something that I would say it’s try to be interested in something else beyond the very basics of the computers themselves.

[I]f you know the basis of computers, if you know about architecture and you know about languages and you know about algorithms, if you know about protocols and databases, it doesn’t really matter if you don’t know the latest thing. And I think one of the big problems in computer science these days is the pressure from students and parents and industry that if you’re not right away able to work with a certain tool, you’re in bad shape. But my advice is don’t worry about that. As long as you know your math, you know your science, you know your whatever, economics or biology, and you know all the basics of computer science, you’ll do very well and you’ll find lots of interesting things to do.

I have a slogan I once invented for myself, that once you learn [verification] you’re a better programmer. And why is this? Because when once you start writing some code, I try to think … suppose somebody pulled a gun on you and said, “Your verification or your life!” Would you be able to verify it? And what I’m saying is that it makes me a more defensive programmer. I don’t try to do clever things because I know that I can get in trouble and I’d never be able to verify it. So maybe I don’t verify everything or almost anything, but it improves the way I do the program.